Blumlein's Binaural Sound (now known as Blumlein Stereo)

Ambisonics (the encoding/decoding solution used in this system) is easier to understand if where it came from is understood. For this reason, Blumlein's Binaural Sound is an important system, and will aid your understanding of coincident microphone techniques.
Alan Blumlein delivered a patent specification in 1931 (Blumlein, 1931) that both recognised the problems with the Bell Labs approach and defined a method for converting spaced microphone feeds to a signal suitable for loudspeaker reproduction. Blumlein called his invention Binaural Reproduction. This recording technique comprised of two omni-directional microphones spaced at a distance similar to that found between the ears, with a round panel baffle in between them. This technique was known to work well for headphone listening, but did not perform as accurately when replayed on loudspeakers. Blumlein_2.GIFBlumlein realised that for loudspeaker reproduction, phase differences at the speakers (i.e. in the spaced microphone recording) did not reproduce phase differences at the listener’s ears. This was due to the unavoidable crosstalk between the two speakers and the two ears of the listener, as shown on the right.

Blumlein had discovered that in order to reproduce phase differences at the ears of a listener, level differences needed to be presented by the speakers. His invention included the description of a ‘Shuffling’ circuit, which is a device that converts the phase differences, present in spaced microphone recordings, to amplitude differences at low frequencies (as at higher frequencies the amplitude differences would already be present due to the sound shadow presented by the disk between the two microphones).

If we consider the stereo pair of loudspeakers shown above, it can be seen that there are two paths from each speaker to each ear of the listener. If the sound that is recorded from the Blumlein stereo pair of microphones is to the left of centre, then the left channel’s signal will be greater in amplitude than the right channel’s signal. Four signals will then be transmitted to the ears:
  1. The left speaker to the left ear.
  2. The left speaker to the right ear.
  3. The right speaker to the right ear.
  4. The right speaker to the left ear.
Blumlein_4.GIFIf we take the case of a low frequency sound (where the interaural phase difference is the major cue), as the paths from the speaker to the contralateral ear is longer than from the speaker to the ipsilateral ear, the signal will appear delayed in time (but not changed in amplitude, due to the wave diffracting around the head). The resulting signals that arrive at each ear are shown in the figure to the right.

It can be clearly seen that low frequency phase cues can be encoded into a stereo signal using just amplitude differences and once the head starts to become a physical obstacle for the reproduced signals (at higher frequencies), a level difference between the ears will also become apparent.

To be continued........

References

Blumlein, A. (1931) Improvements in and relating to Sound-transmission, Sound-recording and Sound-reproducing Systems, British Patent Application 394325.